1/21 🇸🇪 I’ve been looking at cases and ICU admissions lately. In this graphaton 🧵 I’ll present the figures for each region including date of local recommendations - in titel - if there are any.
Stockholm
2/21 Örebro
3/21 Västernorrland
4/21 Värmland
5/21 Uppsala
6/21 Norrbotten
7/21 Kalmar
8/21 Jämtland
9/21 Gävleborg
10/21 Dalarna
11/21 Östergötland
12/21 Västmanland
13/21 Västerbotten
14/21 Västra Götaland
15/21 Sörmland
16/21 Skåne
17/21 Kronoberg
18/21 Jönköping
19/21 Halland
20/21 Gotland
21/21 Blekinge
DM if anything looks odd or I missed something.
Remember, hospitalisations are high but am focusing on ICU since those are the most severe cases.
1/10 🇸🇪 Graphs updated including August. Only September left and this👇🏽 graph is complete. Comparing October through August looking at all cause mortality.
Following tweets will focus on all cause mortality in Sweden:
-No adjustments
-Crude adjustments
-ASMR
-Average
-Trend
+
2/10 Link to my previous thread will be in last tweet, containing more details on Swedish #covid19 data. This thread is shorter and focus on all cause mortality. Here we see January through August. 2020 similar to eg 2012, 2013 & 2015. +
3/10 Looking at rolling three months average we see that current short trend is down since spike in April and that the long trend in general is downwards as well probably due to better general health and better health services. +
1/4 🇸🇪 updating a few graphs, adding one additional week (35)👇🏽 follow the dark green line. At or below average of 2015-2019 for 9 consecutive weeks. Week 35 so far only recorded 10 #covid19 deaths. Will be adjusted upwards but still very few. +
2/4 Year 2020 and year 2019 continues to stay ca 13 days from average of 2015-2018 and when accounting for population growth... +
1/4 🇸🇪 there have been speculation if 2019 with few deaths could impact the outcome of deaths 2020. What does history tell us? Instead of y2y I looked at consecutive flu seasons👇🏽and as we can see deaths increased until 2000 and then slight downward trend until 2014 and then up+
2/4 We see that most drops in all cause mortality (acm) season2season are followed by an increase the year after. Does larger drops (more than -3,37%) result in increases? Yes always during this period. Does the drop revert with equally large percentage increase? +
3/4 Well, so it seems at least. When looking at drops in three ranges we see that often a drop is followed by an on average equally large increase. 2019 had a large drop of acm which was followed by a large increase in 2020.+
1/18 🇸🇪 4th update. Shorter, many aspects are similar to last update, eg age distribution of #covid19 deaths in Sweden -> ca 90% over 70 years old or that ca 70% of those deceased were either care home residents or receiving home care service as can be seen in graph below. +
2/18 Let’s start with two model projections. First the IHME model which shows that prediction of deaths were far above actual. This is continually worth reminding of because these messages were hammered into our heads by media and the #covid19 opinionados. +
3/18 Equally disastrous were the projections of ICU beds needed, hospitals were predicted to be overrun. Very good for us this did not happen and we kept below max capacity at all times. + #COVID19sverige